End of training

README.md

@@ -33,16 +33,15 @@ datasets:
     type: alpaca
     ds_type: json
 
-dataset_prepared_path: noeloco/cameltest
+hf_use_auth_token: true
+push_dataset_to_hub: noeloco
 val_set_size: 0.05
 output_dir: ./lora-out
-
+chat_template: chatml
 hub_model_id: noeloco/modeltest1
-hf_use_auth_token: true
 
 sequence_len: 2048
-sample_packing: true
-eval_sample_packing: False
+sample_packing: false
 pad_to_sequence_len: true
 
 adapter: lora
@@ -61,7 +60,7 @@ wandb_log_model:
 
 gradient_accumulation_steps: 4
 micro_batch_size: 2
-num_epochs: 1
+num_epochs: 2
 optimizer: adamw_bnb_8bit
 lr_scheduler: cosine
 learning_rate: 0.0002
@@ -78,7 +77,7 @@ resume_from_checkpoint:
 local_rank:
 logging_steps: 1
 xformers_attention:
-flash_attention: true
+flash_attention: false
 
 warmup_steps: 10
 evals_per_epoch: 4
@@ -101,7 +100,7 @@ special_tokens:
 
 This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.7345
+- Loss: 0.1202
 
 ## Model description
 
@@ -117,19 +116,6 @@ More information needed
 
 ## Training procedure
 
-
-The following `bitsandbytes` quantization config was used during training:
-- quant_method: bitsandbytes
-- load_in_8bit: True
-- load_in_4bit: False
-- llm_int8_threshold: 6.0
-- llm_int8_skip_modules: None
-- llm_int8_enable_fp32_cpu_offload: False
-- llm_int8_has_fp16_weight: False
-- bnb_4bit_quant_type: fp4
-- bnb_4bit_use_double_quant: False
-- bnb_4bit_compute_dtype: float32
-
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
@@ -142,19 +128,25 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 10
-- num_epochs: 1
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.7285        | 1.0   | 1    | 2.7345          |
+| 1.5644        | 0.06  | 1    | 2.7399          |
+| 1.575         | 0.29  | 5    | 2.6344          |
+| 1.1169        | 0.57  | 10   | 1.2350          |
+| 0.6719        | 0.86  | 15   | 0.5019          |
+| 0.3372        | 1.14  | 20   | 0.2525          |
+| 0.3403        | 1.43  | 25   | 0.1470          |
+| 0.1656        | 1.71  | 30   | 0.1202          |
 
 
 ### Framework versions
 
-- PEFT 0.7.0
-- Transformers 4.37.0.dev0
+- PEFT 0.7.2.dev0
+- Transformers 4.37.0
 - Pytorch 2.0.1+cu118
 - Datasets 2.16.1
 - Tokenizers 0.15.0

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e21d98e65be8ce1f9ac99ceb19b3bd3aa73dd696cc7f7c22965f55c745f9b567
+oid sha256:54e31da3d13f2c089248c59cf1951dfe80272e8d85a2e8ce4cbefb2aaa759ee1
 size 80114765